An Efficient Attack-Resistant Trust Model for P2P Networks

Building trust relationship between participants in a large-scale distributed Peer-to-Peer(P2P) file-sharing system is a challenging research topic because of peer anonymity, peer independence, high dynamics of peer behaviors, and the absence of an effective security mechanism. However, it is difficult to built trust simply by the traditional mechanism. Recommendation based trust models which are similar with and come from social relationship can resolve the problem, but face the challenges of subjectively, experiential weighting recommendation information when aggregating them. This paper presents ARTrust—an Attack Resistant Trust management model, a novel recommendation based trust model for P2P networks. The trust model consists of two parts: reputation evaluation and penalty evaluation. Reputation represents the accumulative assessment of the long-term behavior and the penalty part which is further divided into conflicting value and misusage value is employed to deal with the dynamic or spoiling behavior of peers, which makes ARTrust differ from other trust models based on the reputation only. For the problem of security, some measures are also proposed to defense against several malicious attacks. Subsequent experimental results show that, compared to the existing trust models, our model is more robust on trust security problems and more advanced in successful transaction rate.